Store and retrieve persistent knowledge graph memories via Graphiti MCP server.
npx skills add https://github.com/ships/eden --skill zep-memory
Please help me install this Agent Skill. Skill: zep-memory Source: https://github.com/ships/eden/tree/main/assets/claude/skills/zep-memory Command: npx skills add https://github.com/ships/eden --skill zep-memory
Zep-memory helps AI agents retain important information across sessions by storing and retrieving it from a persistent knowledge graph, so future suggestions and actions respect prior context and discovered preferences.
group_id management (per-user, per-project, per-worktree, etc.).entity_types filtering and targeted fact queries to avoid overly broad results.Use zep-memory to store and later retrieve the user’s project-specific deploy procedure by searching for existing procedures first, then adding the new workflow as a Procedure under a group_id scoped to that project.
High-intent search queries and answers about installing and using this skill.
You can persist agent memory across sessions by using a knowledge graph to store and retrieve preferences, requirements, and procedures, ensuring future actions respect prior context. This approach prevents inconsistency by retaining discovered facts beyond single conversation boundaries.
To prevent context leakage when storing agent knowledge, you must apply explicit group_id scoping to isolate data per user, project, or worktree. Additionally, using targeted entity_types filtering during retrieval prevents knowledge pollution from overly broad search results.
You can save reusable procedures and requirements by using the add_memory operation with appropriate source types like text, json, or message. This immediately captures new workflows and facts into the persistent graph as soon as they are discovered during interactions.
Yes, you can retrieve specific entity facts without returning irrelevant data by applying explicit entity_types filtering during your search operations. This entity-focused, anti-bloat retrieval mechanism encourages targeted fact queries to avoid overly broad results.
Yes, agent memory works with the Graphiti MCP server to provide persistent knowledge graph storage and retrieval. This integration enables cross-session retention of preferences and reusable procedures across different conversation boundaries and project contexts.
daffy0208Design, implement, and maintain knowledge graphs for complex entity relationships.
Toperythroblast876Search the ourmem shared memory store for relevant stored memories via semantic search.
arunrlvermaStore and retrieve AI agent facts, decisions, and preferences in a cloud JSON file.
rahulgupta2018Designs persistent memory architectures for agent systems with cross-session retention and entity tracking.